From ebad78163b34e53786338ac8778c71c02cd090a3 Mon Sep 17 00:00:00 2001 From: "Niall Douglas (s [underscore] sourceforge {at} nedprod [dot] com)" Date: Sun, 27 Aug 2017 05:00:10 +0100 Subject: Update storage profile database with latest results. --- release_notes.md | 17 ++++++++++------- 1 file changed, 10 insertions(+), 7 deletions(-) (limited to 'release_notes.md') diff --git a/release_notes.md b/release_notes.md index 1ce08286..80978374 100644 --- a/release_notes.md +++ b/release_notes.md @@ -21,32 +21,34 @@ source code repository lives at https://github.com/ned14/boost.afio. Manufacturer claimed 4Kb transfer latencies for the physical hardware: -- Spinning rust hard drive latency @ QD1: **7000us** +- Spinning rust hard drive latency @ QD1: **9000us** - SATA flash drive latency @ QD1: **800us** - NVMe flash drive latency @ QD1: **300us** - RTT UDP packet latency over a LAN: **60us** -- NVMe Optane drive latency @ QD1: **10us** +- NVMe Optane drive latency @ QD1: **60us** - `memcpy(4Kb)` latency: **5us** (main memory) to **1.3us** (L3 cache) - RTT PCIe latency: **0.5us** 100% read QD1 4Kb direct transfer latencies for the software with AFIO: -- < 99% spinning rust hard drive latency: **230,005us** +- < 99% spinning rust hard drive latency: Windows **187,231us** FreeBSD **9,836us** Linux **26,468us** - < 99% SATA flash drive latency: Windows **290us** Linux **158us** +- < 99% NVMe drive latency: Windows **150us** FreeBSD **70us** Linux **30us** -Lowest sustained 4Kb read latency benchmarked to date by author (page cache): **7us** (142k IOPS @ QD1, approx 558Mb/sec) +Lowest achievable 4Kb read latency benchmarked to date by author: **0.19us** (5.3M IOPS @ QD1, approx 20Gb/sec) 75% read 25% write QD4 4Kb direct transfer latencies for the software with AFIO: -- < 99% spinning rust hard drive latency: **48,185us** +- < 99% spinning rust hard drive latency: Windows **48,185us** FreeBSD **61,834us** Linux **104,507us** - < 99% SATA flash drive latency: Windows **1,812us** Linux **1,416us** +- < 99% NVMe drive latency: Windows **95us** FreeBSD **143us** Linux **40us** -Lowest sustained 4Kb write latency benchmarked to date by author (page cache): **7us** (142k IOPS @ QD1, approx 558Mb/sec) +Lowest achievable 4Kb write latency benchmarked to date by author: **0.18us** (5.6M IOPS @ QD1, approx 21Gb/sec) -\note Note that this code is of late alpha quality. It's quite reliable, but be careful when using it! +\note Note that this code is of late alpha quality. It's quite reliable on Windows and Linux, but be careful when using it! These compilers and OS are regularly tested: - GCC 7.0 (Linux 4,x x64) @@ -146,6 +148,7 @@ Todo to reach feature parity with AFIO v1: | NEW in v2 | Windows | POSIX | | | --------- | --------| ----- | --- | | | | | Hard links and symlinks. +| | | | Set random or sequential i/o (prefetch). | | | | BSD and OS X kqueues optimised `io_service` Todo thereafter: -- cgit v1.2.3